An epidemiological study of the records of the 2,000-odd cases of children's cancer included in the Manchester Children's Tumour Registry since January 1, 1954, in order to search for aetiological factors. This registry is strictly population based and every effort has been made to ensure diagnostic accuracy and full coverage of all childrn under 15 years old who develop tumours and who live inthe Manchester Hospital Region. (Population under 15 years, 1 million). The study would be aimed at examining the following factors: (a) The incidence of the tumours of each type and at each site by age of onset. (b) Variations in incidence with space, time and space-time. (c) Variations in incidence with personal and family circumstances including incidents during the relevant pegnancy. (d) Associated pathology in the children and their relatives. (e) The health of long-term survivors and their progeny.